用于分类，找出新数据的最近邻的相似数据，按照近邻数据的分类来分类。(Similar data used to classify and locate nearest neighbors of new data, classified according to the classification of nearest neighbor data.)